Despite the decomposition of convolutional kernels for lightweight CNNs being well studied, existing works that rely on tensor network diagrams or hyperdimensional abstraction lack geometry intuition. This work devises a new perspective by linking a 3D-reshaped kernel tensor to its various slice-wise and rank-1 decompositions, permitting a straightforward connection between various tensor approximations and efficient CNN modules. Specifically, it is discovered that a pointwise-depthwise-pointwise (PDP) configuration constitutes a viable construct for lightweight CNNs. Moreover, a novel link to the latest ShiftNet is established, inspiring a first-ever shift layer pruning that achieves nearly 50% compression with < 1% drop in accuracy for ShiftResNet.

翻译：尽管用于轻量级CNN的卷积核分解已被广泛研究，但现有依赖张量网络图或超维抽象的方法缺乏几何直观性。本文通过将三维重塑后的核张量与其各类切片分解和秩-1分解相关联，提出了一种新视角，从而在多种张量近似方法与高效CNN模块之间建立直接联系。具体而言，研究发现逐点-深度-逐点（PDP）结构是构建轻量级CNN的可行方案。此外，本文还首次建立了与最新ShiftNet之间的联系，并由此启发了首个移位层剪枝方法——该方法在ShiftResNet上实现了近50%的压缩率，同时准确率下降不足1%。